Orthogonal frequency division multiplexing (OFDM) has been adopted as standard beginning with the 4thgeneration mobile communication system because of its high-bit-rate transmission capability under...
1 January 2018 in International Journal of Computer Theory and Engineering
Edge computing brings computational ability to network edges to enable low latency based on deploying devices close to the environment where the data is generated. Nevertheless, the limitation of size and energy consumption constrain the scalability and performance of edge device applications such as deep learning, although, cloud computing can be adopted to support high-performance tasks with centralized data collection. However, frequently communicating with a central cloud server brings potential risks to security and privacy issues by exposing data on the Internet. In this paper, we propose a secure continuous knowledge transfer approach to improve knowledge by collaborating with multiple edge devices in the decentralized edge computing architecture without a central server. Using blockchain, the knowledge integrity is maintained in the transfer process by recording the transaction information of each knowledge improvement and synchronizing the blockchain in each edge device. The knowledge is a trained deep-learning model that is derived by learning the local data. Using the local data of each edge device, the model is continuously trained to improve performance. Therefore, each improvement is recorded as the contribution of each edge device immutably in the decentralized edge computing architecture.
Accurate acoustic identification of marine mammals is vital for monitoring ocean health and human impacts. Existing methods often struggle with limited single-feature representations or suboptimal fusion of multiple features. This paper proposes an Evaluation-Adaptive Weighted Multi-Head Fusion Network that integrates CQT and STFT features via a dual-branch ResNet architecture. The model enhances intra-branch features using channel attention and adaptive weighting of each branch based on its validation accuracy during training. Experiments on the Watkins Marine Mammal Sound Database show that the proposed method achieves superior performance, reaching 96.05% accuracy and outperforming baseline and attention-based fusion models. This approach offers an effective solution for multi-feature acoustic recognition in complex underwater environments.
Underwater wireless sensor networks play an important role in underwater communication systems. Communication through collaborative communication is an effective way to solve critical problems in underwater communication systems. Underwater sensors are often deployed in spaces that overlap with those of marine mammals, which can adversely affect them. For this reason, in this paper, a marine mammal conflict avoidance method that can be dynamically adjusted according to the channel idle time duration and sensor node demand is designed, and the derivation of the maximum occupancy time duration is performed. Meanwhile, in addition, combining the potential of reinforcement learning in adaptive management, efficient resource optimization, and solving complex problems, this study also proposes a reinforcement learning-based relay-assisted spectrum switching method (R2S), which aims to achieve a reasonable allocation of spectrum resources in relay collaborative communication systems. The experimental results show that the method proposed in this study can effectively reduce the disturbance to marine mammals while performing well in terms of conflict probability, interruption probability, and quality of service.
FlexRay is becoming the in-vehicle communication network of the next generation. In this study, the main contents are the FlexRay network static segment scheduling algorithm and optimization strategy, improve the scheduling efficiency of vehicle network and optimize the performance of communication network. The FlexRay static segment characteristic was first analyzed, then selected bandwidth utilization as the performance metrics to scheduling problem. A signal packing method is proposed based on Next Fit Decreasing (NFD) algorithm. Then Frame ID (FID) multiplexing method was used to minimize the number of FIDs. Finally, experimental simulation by CANoe. FlexRay software, that shows the model can quickly obtain the message schedule of each node, effectively control the message payload size and reduced bus payload by 16.3%, the number of FID drops 53.8% while improving bandwidth utilization by 32.8%.